Cookies

We use cookies to ensure that we give you the best experience on our website. You can change your cookie settings at any time. Otherwise, we'll assume you're OK to continue.

Department of Mathematical Sciences

Seminar Archives

On this page you can find information about seminars in this and previous academic years, where available on the database.

Statistics Seminars: An extensible software platform for the development and evaluation of Machine Learning methods

Presented by Andres Masegosa, University of Granada

31 August 2009 14:15 in CM107

The implementation and evaluation of new machine-learning methods is a cumbersome task. New machine-learning proposals should be evaluated with a wide range of benchmark datasets and compare against some of the state-of-the-art algorithms in order to empirically show the performance of new proposed methods. This task can become a burdensome and time-consuming task if we have to design, develop and test all data structures and algorithms which are needed to implement and evaluate any machine-learning method. Time and memory efficiency issues often arise when this software is developed.

Weka is an open-source java software equipped with a wide range of facilities that ease the implementation and evaluation of new machine-learning methods. In this talk, the main features of this highly cited software will be highlighted. Moreover, it will be detailed a case study with the implementation and evaluation of simple credal decision trees.

Contact sunil.chhita@durham.ac.uk for more information